Temporal Link Prediction: A Unified Framework, Taxonomy, and Review

Abstract

Dynamic graphs serve as a generic abstraction and description of the evolutionary behaviors of various complex systems (e.g., social networks and communication networks). Temporal link prediction (TLP) is a classic inference task on dynamic graphs, which aims to predict possible future linkage using historical dynamic topology. The predicted future topology can be used to support some advanced applications on real-world systems (e.g., resource pre-allocation) for better system performance. This survey provides a comprehensive review of existing representative TLP methods. Concretely, we first give the formal statements regarding data models, task settings, and learning paradigms that are commonly used in related research. A hierarchical fine-grained taxonomy is further introduced to categorize existing methods in terms of their data models, learning paradigms, and techniques. From a generic perspective, we propose a unified encoder-decoder framework to formulate all the methods reviewed, where different approaches only differ in terms of some components of the unified framework. Moreover, we envision serving the community with an open-source project OpenTLP 1 that refactors or implements some representative TLP methods using the proposed unified framework and summarizes other public sources. To conclude this survey, we also discuss some advanced topics in recent research and highlight possible future directions. CCS Concepts: • General and reference → Surveys and overviews; • Networks → Network dynamics; • Information systems → Temporal data.

Graph Neural Networks for temporal graphs: State of the art, open challenges, and opportunities

“…Thus, we consider the setting to be transductive when both node instances are known at training time, and inductive otherwise. Instead, [33] adopt a different approach and identify Level-1 (the set of nodes is fixed) and Level-2 (nodes may be added and removed over time) temporal link prediction tasks.…”

“…However, despite the potential of GNN-based models for temporal graph processing and the variety of different approaches that emerged, a systematization of the literature is still missing. Existing surveys either discuss general techniques for learning over temporal graphs, only briefly mentioning temporal extensions of GNNs [20,2,52,49], or focus on specific topics, like temporal link prediction [33,41] or temporal graph generation [15]. This work aims to fill this gap by providing a systematization of existing GNN-based methods for temporal graphs, or Temporal GNNs (TGNNs), and a formalization of the tasks being addressed.…”